[FFmpeg-devel] [PATCH] Fix configure to disable MMX for Pentium Pro

Måns Rullgård mans
Thu Aug 26 15:47:37 CEST 2010


Alex Converse <alex.converse at gmail.com> writes:

> 2010/8/26 M?ns Rullg?rd <mans at mansr.com>:
>> Alex Converse <alex.converse at gmail.com> writes:
>>
>>> 2010/8/26 M?ns Rullg?rd <mans at mansr.com>:
>>>> Janne Grunau <janne-ffmpeg at jannau.net> writes:
>>>>
>>>>> On Thu, Aug 26, 2010 at 01:52:22AM -0400, Jason Garrett-Glaser wrote:
>>>>>> On Thu, Aug 26, 2010 at 12:44 AM, Eli Friedman <eli.friedman at gmail.com> wrote:
>>>>>> > Attached. ?This is essentially fixing an oversight in r21277.
>>>>>> >
>>>>>> > -Eli
>>>>>>
>>>>>> This means every distro who uses --cpu=i686 or similar will silently
>>>>>> have their ffmpeg crippled the next time they make a build.
>>>>>>
>>>>>> I would rather roll back r21277 than apply this patch.
>>>>>
>>>>> What about requriring --enable-runtime-cpudetect for i686 with mmx and
>>>>> crippling emms() in that case like attached patch?
>>>>
>>>> I'm not changing a thing until someone shows me evidence of ffmpeg
>>>> actually being used on such CPUs.
>>>
>>> We shouldn't claim to support something that doesn't work.
>>
>> But we _do_ support them. ?Disable mmx and it will run on a 486 even.
>>
>
> ./configure --cpu=prentiumpro says build me an FFmpeg for a
> PentiumPro. The result will no longer run on a PentiumPro.
>
> That is broken behavior. It doesn't matter that the chip is old.

Fixed.

-- 
M?ns Rullg?rd
mans at mansr.com



More information about the ffmpeg-devel mailing list